X-Git-Url: http://git.rot13.org/?a=blobdiff_plain;f=fs%2Fgfs2%2Fops_address.c;h=d8d69a72a10dd819bfcdbe6a736009bda5a92e27;hb=eb991b39385c7b04923d701764a34694ec54b53d;hp=207363aed11216dd7da057e99df07cfbb1142a84;hpb=feaa7bba026c181ce071d5a4884f7f9dd26207a1;p=powerpc.git diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index 207363aed1..d8d69a72a1 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -4,7 +4,7 @@ * * This copyrighted material is made available to anyone wishing to use, * modify, copy, or redistribute it subject to the terms and conditions - * of the GNU General Public License v.2. + * of the GNU General Public License version 2. */ #include @@ -17,9 +17,9 @@ #include #include #include +#include #include "gfs2.h" -#include "lm_interface.h" #include "incore.h" #include "bmap.h" #include "glock.h" @@ -27,12 +27,30 @@ #include "log.h" #include "meta_io.h" #include "ops_address.h" -#include "page.h" #include "quota.h" #include "trans.h" #include "rgrp.h" #include "ops_file.h" #include "util.h" +#include "glops.h" + + +static void gfs2_page_add_databufs(struct gfs2_inode *ip, struct page *page, + unsigned int from, unsigned int to) +{ + struct buffer_head *head = page_buffers(page); + unsigned int bsize = head->b_size; + struct buffer_head *bh; + unsigned int start, end; + + for (bh = head, start = 0; bh != head || !start; + bh = bh->b_this_page, start = end) { + end = start + bsize; + if (end <= from || start >= to) + continue; + gfs2_trans_add_bh(ip->i_gl, bh, 0); + } +} /** * gfs2_get_block - Fills in a buffer head with details about a block @@ -47,29 +65,11 @@ int gfs2_get_block(struct inode *inode, sector_t lblock, struct buffer_head *bh_result, int create) { - int new = create; - uint64_t dblock; - int error; - int boundary; - - error = gfs2_block_map(inode, lblock, &new, &dblock, &boundary); - if (error) - return error; - - if (!dblock) - return 0; - - map_bh(bh_result, inode->i_sb, dblock); - if (new) - set_buffer_new(bh_result); - if (boundary) - set_buffer_boundary(bh_result); - - return 0; + return gfs2_block_map(inode, lblock, create, bh_result); } /** - * get_block_noalloc - Fills in a buffer head with details about a block + * gfs2_get_block_noalloc - Fills in a buffer head with details about a block * @inode: The inode * @lblock: The block number to look up * @bh_result: The buffer head to return the result in @@ -78,26 +78,23 @@ int gfs2_get_block(struct inode *inode, sector_t lblock, * Returns: errno */ -static int get_block_noalloc(struct inode *inode, sector_t lblock, - struct buffer_head *bh_result, int create) +static int gfs2_get_block_noalloc(struct inode *inode, sector_t lblock, + struct buffer_head *bh_result, int create) { - int new = 0; - uint64_t dblock; int error; - int boundary; - error = gfs2_block_map(inode, lblock, &new, &dblock, &boundary); + error = gfs2_block_map(inode, lblock, 0, bh_result); if (error) return error; + if (bh_result->b_blocknr == 0) + return -EIO; + return 0; +} - if (dblock) - map_bh(bh_result, inode->i_sb, dblock); - else if (gfs2_assert_withdraw(GFS2_SB(inode), !create)) - error = -EIO; - if (boundary) - set_buffer_boundary(bh_result); - - return error; +static int gfs2_get_block_direct(struct inode *inode, sector_t lblock, + struct buffer_head *bh_result, int create) +{ + return gfs2_block_map(inode, lblock, 0, bh_result); } /** @@ -113,8 +110,8 @@ static int get_block_noalloc(struct inode *inode, sector_t lblock, static int gfs2_writepage(struct page *page, struct writeback_control *wbc) { struct inode *inode = page->mapping->host; - struct gfs2_inode *ip = GFS2_I(page->mapping->host); - struct gfs2_sbd *sdp = GFS2_SB(page->mapping->host); + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_sbd *sdp = GFS2_SB(inode); loff_t i_size = i_size_read(inode); pgoff_t end_index = i_size >> PAGE_CACHE_SHIFT; unsigned offset; @@ -140,10 +137,14 @@ static int gfs2_writepage(struct page *page, struct writeback_control *wbc) error = gfs2_trans_begin(sdp, RES_DINODE + 1, 0); if (error) goto out_ignore; + if (!page_has_buffers(page)) { + create_empty_buffers(page, inode->i_sb->s_blocksize, + (1 << BH_Dirty)|(1 << BH_Uptodate)); + } gfs2_page_add_databufs(ip, page, 0, sdp->sd_vfs->s_blocksize-1); done_trans = 1; } - error = block_write_full_page(page, get_block_noalloc, wbc); + error = block_write_full_page(page, gfs2_get_block_noalloc, wbc); if (done_trans) gfs2_trans_end(sdp); gfs2_meta_cache_flush(ip); @@ -155,19 +156,6 @@ out_ignore: return 0; } -static int zero_readpage(struct page *page) -{ - void *kaddr; - - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr, 0, PAGE_CACHE_SIZE); - kunmap_atomic(page, KM_USER0); - - SetPageUptodate(page); - - return 0; -} - /** * stuffed_readpage - Fill in a Linux page with stuffed file data * @ip: the inode @@ -182,9 +170,7 @@ static int stuffed_readpage(struct gfs2_inode *ip, struct page *page) void *kaddr; int error; - /* Only the first page of a stuffed file might contain data */ - if (unlikely(page->index)) - return zero_readpage(page); + BUG_ON(page->index); error = gfs2_meta_inode_buffer(ip, &dibh); if (error) @@ -194,7 +180,7 @@ static int stuffed_readpage(struct gfs2_inode *ip, struct page *page) memcpy(kaddr, dibh->b_data + sizeof(struct gfs2_dinode), ip->i_di.di_size); memset(kaddr + ip->i_di.di_size, 0, PAGE_CACHE_SIZE - ip->i_di.di_size); - kunmap_atomic(page, KM_USER0); + kunmap_atomic(kaddr, KM_USER0); brelse(dibh); @@ -217,16 +203,26 @@ static int gfs2_readpage(struct file *file, struct page *page) { struct gfs2_inode *ip = GFS2_I(page->mapping->host); struct gfs2_sbd *sdp = GFS2_SB(page->mapping->host); + struct gfs2_file *gf = NULL; struct gfs2_holder gh; int error; - - if (likely(file != &gfs2_internal_file_sentinal)) { - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME|GL_AOP, &gh); - error = gfs2_glock_nq_m_atime(1, &gh); + int do_unlock = 0; + + if (likely(file != &gfs2_internal_file_sentinel)) { + if (file) { + gf = file->private_data; + if (test_bit(GFF_EXLOCK, &gf->f_flags)) + /* gfs2_sharewrite_nopage has grabbed the ip->i_gl already */ + goto skip_lock; + } + gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME|LM_FLAG_TRY_1CB, &gh); + do_unlock = 1; + error = gfs2_glock_nq_atime(&gh); if (unlikely(error)) goto out_unlock; } +skip_lock: if (gfs2_is_stuffed(ip)) { error = stuffed_readpage(ip, page); unlock_page(page); @@ -236,21 +232,21 @@ static int gfs2_readpage(struct file *file, struct page *page) if (unlikely(test_bit(SDF_SHUTDOWN, &sdp->sd_flags))) error = -EIO; - if (file != &gfs2_internal_file_sentinal) { + if (do_unlock) { gfs2_glock_dq_m(1, &gh); gfs2_holder_uninit(&gh); } out: return error; out_unlock: + if (error == GLR_TRYFAILED) + error = AOP_TRUNCATED_PAGE; unlock_page(page); - if (file != &gfs2_internal_file_sentinal) + if (do_unlock) gfs2_holder_uninit(&gh); goto out; } -#define list_to_page(head) (list_entry((head)->prev, struct page, lru)) - /** * gfs2_readpages - Read a bunch of pages at once * @@ -275,22 +271,30 @@ static int gfs2_readpages(struct file *file, struct address_space *mapping, struct gfs2_holder gh; unsigned page_idx; int ret; + int do_unlock = 0; - if (likely(file != &gfs2_internal_file_sentinal)) { + if (likely(file != &gfs2_internal_file_sentinel)) { + if (file) { + struct gfs2_file *gf = file->private_data; + if (test_bit(GFF_EXLOCK, &gf->f_flags)) + goto skip_lock; + } gfs2_holder_init(ip->i_gl, LM_ST_SHARED, - LM_FLAG_TRY_1CB|GL_ATIME|GL_AOP, &gh); - ret = gfs2_glock_nq_m_atime(1, &gh); - if (ret == GLR_TRYFAILED) + LM_FLAG_TRY_1CB|GL_ATIME, &gh); + do_unlock = 1; + ret = gfs2_glock_nq_atime(&gh); + if (ret == GLR_TRYFAILED) goto out_noerror; if (unlikely(ret)) goto out_unlock; } - +skip_lock: if (gfs2_is_stuffed(ip)) { struct pagevec lru_pvec; pagevec_init(&lru_pvec, 0); for (page_idx = 0; page_idx < nr_pages; page_idx++) { - struct page *page = list_to_page(pages); + struct page *page = list_entry(pages->prev, struct page, lru); + prefetchw(&page->flags); list_del(&page->lru); if (!add_to_page_cache(page, mapping, page->index, GFP_KERNEL)) { @@ -298,8 +302,9 @@ static int gfs2_readpages(struct file *file, struct address_space *mapping, unlock_page(page); if (!pagevec_add(&lru_pvec, page)) __pagevec_lru_add(&lru_pvec); + } else { + page_cache_release(page); } - page_cache_release(page); } pagevec_lru_add(&lru_pvec); ret = 0; @@ -308,7 +313,7 @@ static int gfs2_readpages(struct file *file, struct address_space *mapping, ret = mpage_readpages(mapping, pages, nr_pages, gfs2_get_block); } - if (likely(file != &gfs2_internal_file_sentinal)) { + if (do_unlock) { gfs2_glock_dq_m(1, &gh); gfs2_holder_uninit(&gh); } @@ -319,14 +324,7 @@ out: out_noerror: ret = 0; out_unlock: - /* unlock all pages, we can't do any I/O right now */ - for (page_idx = 0; page_idx < nr_pages; page_idx++) { - struct page *page = list_to_page(pages); - list_del(&page->lru); - unlock_page(page); - page_cache_release(page); - } - if (likely(file != &gfs2_internal_file_sentinal)) + if (do_unlock) gfs2_holder_uninit(&gh); goto out; } @@ -352,19 +350,25 @@ static int gfs2_prepare_write(struct file *file, struct page *page, loff_t pos = ((loff_t)page->index << PAGE_CACHE_SHIFT) + from; loff_t end = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to; struct gfs2_alloc *al; + unsigned int write_len = to - from; - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_ATIME|GL_AOP, &ip->i_gh); - error = gfs2_glock_nq_m_atime(1, &ip->i_gh); - if (error) + + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_ATIME|LM_FLAG_TRY_1CB, &ip->i_gh); + error = gfs2_glock_nq_atime(&ip->i_gh); + if (unlikely(error)) { + if (error == GLR_TRYFAILED) + error = AOP_TRUNCATED_PAGE; goto out_uninit; + } - gfs2_write_calc_reserv(ip, to - from, &data_blocks, &ind_blocks); + gfs2_write_calc_reserv(ip, write_len, &data_blocks, &ind_blocks); - error = gfs2_write_alloc_required(ip, pos, from - to, &alloc_required); + error = gfs2_write_alloc_required(ip, pos, write_len, &alloc_required); if (error) goto out_unlock; + ip->i_alloc.al_requested = 0; if (alloc_required) { al = gfs2_alloc_get(ip); @@ -372,7 +376,7 @@ static int gfs2_prepare_write(struct file *file, struct page *page, if (error) goto out_alloc_put; - error = gfs2_quota_check(ip, ip->i_di.di_uid, ip->i_di.di_gid); + error = gfs2_quota_check(ip, ip->i_inode.i_uid, ip->i_inode.i_gid); if (error) goto out_qunlock; @@ -394,8 +398,7 @@ static int gfs2_prepare_write(struct file *file, struct page *page, if (gfs2_is_stuffed(ip)) { if (end > sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode)) { - error = gfs2_unstuff_dinode(ip, gfs2_unstuffer_page, - page); + error = gfs2_unstuff_dinode(ip, page); if (error == 0) goto prepare_write; } else if (!PageUptodate(page)) @@ -443,7 +446,8 @@ static int gfs2_commit_write(struct file *file, struct page *page, struct gfs2_sbd *sdp = GFS2_SB(inode); int error = -EOPNOTSUPP; struct buffer_head *dibh; - struct gfs2_alloc *al = &ip->i_alloc;; + struct gfs2_alloc *al = &ip->i_alloc; + struct gfs2_dinode *di; if (gfs2_assert_withdraw(sdp, gfs2_glock_is_locked_by_me(ip->i_gl))) goto fail_nounlock; @@ -453,22 +457,25 @@ static int gfs2_commit_write(struct file *file, struct page *page, goto fail_endtrans; gfs2_trans_add_bh(ip->i_gl, dibh, 1); + di = (struct gfs2_dinode *)dibh->b_data; if (gfs2_is_stuffed(ip)) { - uint64_t file_size; + u64 file_size; void *kaddr; - file_size = ((uint64_t)page->index << PAGE_CACHE_SHIFT) + to; + file_size = ((u64)page->index << PAGE_CACHE_SHIFT) + to; kaddr = kmap_atomic(page, KM_USER0); memcpy(dibh->b_data + sizeof(struct gfs2_dinode) + from, - (char *)kaddr + from, to - from); - kunmap_atomic(page, KM_USER0); + kaddr + from, to - from); + kunmap_atomic(kaddr, KM_USER0); SetPageUptodate(page); - if (inode->i_size < file_size) + if (inode->i_size < file_size) { i_size_write(inode, file_size); + mark_inode_dirty(inode); + } } else { if (sdp->sd_args.ar_data == GFS2_DATA_ORDERED || gfs2_is_jdata(ip)) @@ -478,10 +485,11 @@ static int gfs2_commit_write(struct file *file, struct page *page, goto fail; } - if (ip->i_di.di_size < inode->i_size) + if (ip->i_di.di_size < inode->i_size) { ip->i_di.di_size = inode->i_size; + di->di_size = cpu_to_be64(inode->i_size); + } - gfs2_dinode_out(&ip->i_di, dibh->b_data); brelse(dibh); gfs2_trans_end(sdp); if (al->al_requested) { @@ -545,10 +553,8 @@ static void discard_buffer(struct gfs2_sbd *sdp, struct buffer_head *bh) if (bd) { bd->bd_bh = NULL; bh->b_private = NULL; - gfs2_log_unlock(sdp); - brelse(bh); - } else - gfs2_log_unlock(sdp); + } + gfs2_log_unlock(sdp); lock_buffer(bh); clear_buffer_dirty(bh); @@ -562,7 +568,7 @@ static void discard_buffer(struct gfs2_sbd *sdp, struct buffer_head *bh) static void gfs2_invalidatepage(struct page *page, unsigned long offset) { - struct gfs2_sbd *sdp = page->mapping->host->i_sb->s_fs_info; + struct gfs2_sbd *sdp = GFS2_SB(page->mapping->host); struct buffer_head *head, *bh, *next; unsigned int curr_off = 0; @@ -588,8 +594,9 @@ static void gfs2_invalidatepage(struct page *page, unsigned long offset) return; } -static ssize_t gfs2_direct_IO_write(struct kiocb *iocb, const struct iovec *iov, - loff_t offset, unsigned long nr_segs) +static ssize_t gfs2_direct_IO(int rw, struct kiocb *iocb, + const struct iovec *iov, loff_t offset, + unsigned long nr_segs) { struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; @@ -597,15 +604,20 @@ static ssize_t gfs2_direct_IO_write(struct kiocb *iocb, const struct iovec *iov, struct gfs2_holder gh; int rv; + if (rw == READ) + mutex_lock(&inode->i_mutex); /* - * Shared lock, even though its write, since we do no allocation + * Shared lock, even if its a write, since we do no allocation * on this path. All we need change is atime. */ gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME, &gh); - rv = gfs2_glock_nq_m_atime(1, &gh); + rv = gfs2_glock_nq_atime(&gh); if (rv) goto out; + if (offset > i_size_read(inode)) + goto out; + /* * Should we return an error here? I can't see that O_DIRECT for * a journaled file makes any sense. For now we'll silently fall @@ -618,44 +630,138 @@ static ssize_t gfs2_direct_IO_write(struct kiocb *iocb, const struct iovec *iov, if (gfs2_is_stuffed(ip)) goto out; - rv = __blockdev_direct_IO(WRITE, iocb, inode, inode->i_sb->s_bdev, - iov, offset, nr_segs, gfs2_get_block, - NULL, DIO_OWN_LOCKING); + rv = blockdev_direct_IO_own_locking(rw, iocb, inode, + inode->i_sb->s_bdev, + iov, offset, nr_segs, + gfs2_get_block_direct, NULL); out: gfs2_glock_dq_m(1, &gh); gfs2_holder_uninit(&gh); + if (rw == READ) + mutex_unlock(&inode->i_mutex); return rv; } /** - * gfs2_direct_IO + * stuck_releasepage - We're stuck in gfs2_releasepage(). Print stuff out. + * @bh: the buffer we're stuck on * - * This is called with a shared lock already held for the read path. - * Currently, no locks are held when the write path is called. */ -static ssize_t gfs2_direct_IO(int rw, struct kiocb *iocb, - const struct iovec *iov, loff_t offset, - unsigned long nr_segs) + +static void stuck_releasepage(struct buffer_head *bh) { - struct file *file = iocb->ki_filp; - struct inode *inode = file->f_mapping->host; - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_sbd *sdp = GFS2_SB(inode); + struct inode *inode = bh->b_page->mapping->host; + struct gfs2_sbd *sdp = inode->i_sb->s_fs_info; + struct gfs2_bufdata *bd = bh->b_private; + struct gfs2_glock *gl; +static unsigned limit = 0; + + if (limit > 3) + return; + limit++; + + fs_warn(sdp, "stuck in gfs2_releasepage() %p\n", inode); + fs_warn(sdp, "blkno = %llu, bh->b_count = %d\n", + (unsigned long long)bh->b_blocknr, atomic_read(&bh->b_count)); + fs_warn(sdp, "pinned = %u\n", buffer_pinned(bh)); + fs_warn(sdp, "bh->b_private = %s\n", (bd) ? "!NULL" : "NULL"); + + if (!bd) + return; - if (rw == WRITE) - return gfs2_direct_IO_write(iocb, iov, offset, nr_segs); + gl = bd->bd_gl; - if (gfs2_assert_warn(sdp, gfs2_glock_is_locked_by_me(ip->i_gl)) || - gfs2_assert_warn(sdp, !gfs2_is_stuffed(ip))) - return -EINVAL; + fs_warn(sdp, "gl = (%u, %llu)\n", + gl->gl_name.ln_type, (unsigned long long)gl->gl_name.ln_number); - return __blockdev_direct_IO(READ, iocb, inode, inode->i_sb->s_bdev, iov, - offset, nr_segs, gfs2_get_block, NULL, - DIO_OWN_LOCKING); + fs_warn(sdp, "bd_list_tr = %s, bd_le.le_list = %s\n", + (list_empty(&bd->bd_list_tr)) ? "no" : "yes", + (list_empty(&bd->bd_le.le_list)) ? "no" : "yes"); + + if (gl->gl_ops == &gfs2_inode_glops) { + struct gfs2_inode *ip = gl->gl_object; + unsigned int x; + + if (!ip) + return; + + fs_warn(sdp, "ip = %llu %llu\n", + (unsigned long long)ip->i_num.no_formal_ino, + (unsigned long long)ip->i_num.no_addr); + + for (x = 0; x < GFS2_MAX_META_HEIGHT; x++) + fs_warn(sdp, "ip->i_cache[%u] = %s\n", + x, (ip->i_cache[x]) ? "!NULL" : "NULL"); + } +} + +/** + * gfs2_releasepage - free the metadata associated with a page + * @page: the page that's being released + * @gfp_mask: passed from Linux VFS, ignored by us + * + * Call try_to_free_buffers() if the buffers in this page can be + * released. + * + * Returns: 0 + */ + +int gfs2_releasepage(struct page *page, gfp_t gfp_mask) +{ + struct inode *aspace = page->mapping->host; + struct gfs2_sbd *sdp = aspace->i_sb->s_fs_info; + struct buffer_head *bh, *head; + struct gfs2_bufdata *bd; + unsigned long t = jiffies + gfs2_tune_get(sdp, gt_stall_secs) * HZ; + + if (!page_has_buffers(page)) + goto out; + + head = bh = page_buffers(page); + do { + while (atomic_read(&bh->b_count)) { + if (!atomic_read(&aspace->i_writecount)) + return 0; + + if (!(gfp_mask & __GFP_WAIT)) + return 0; + + if (time_after_eq(jiffies, t)) { + stuck_releasepage(bh); + /* should we withdraw here? */ + return 0; + } + + yield(); + } + + gfs2_assert_warn(sdp, !buffer_pinned(bh)); + gfs2_assert_warn(sdp, !buffer_dirty(bh)); + + gfs2_log_lock(sdp); + bd = bh->b_private; + if (bd) { + gfs2_assert_warn(sdp, bd->bd_bh == bh); + gfs2_assert_warn(sdp, list_empty(&bd->bd_list_tr)); + gfs2_assert_warn(sdp, !bd->bd_ail); + bd->bd_bh = NULL; + if (!list_empty(&bd->bd_le.le_list)) + bd = NULL; + bh->b_private = NULL; + } + gfs2_log_unlock(sdp); + if (bd) + kmem_cache_free(gfs2_bufdata_cachep, bd); + + bh = bh->b_this_page; + } while (bh != head); + +out: + return try_to_free_buffers(page); } -struct address_space_operations gfs2_file_aops = { +const struct address_space_operations gfs2_file_aops = { .writepage = gfs2_writepage, .readpage = gfs2_readpage, .readpages = gfs2_readpages, @@ -664,6 +770,7 @@ struct address_space_operations gfs2_file_aops = { .commit_write = gfs2_commit_write, .bmap = gfs2_bmap, .invalidatepage = gfs2_invalidatepage, + .releasepage = gfs2_releasepage, .direct_IO = gfs2_direct_IO, };